Loading...

Unsupervised Command Detection in EEG-based Brain-computer Interface

Behmand, Arash | 2017

572 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 49617 (19)
  4. University: Sharif University of Technology
  5. Department: Computer Engineering
  6. Advisor(s): Beigy, Hamid
  7. Abstract:
  8. A Brain–Computer Interface is a system that provides a direct pathway for communication between a brain and a computer device by processing signals from sensors measuring brain activity (here Electroencephalography signals). Brain signals are known to be stochastic, non-stationary, non-linear and highly noisy, Therfore Brain–Computer Interface Systems rely on signal preprocessing, feature extraction and use of machine learning methods in order to detect mental state of Brain–Computer Interface user. Current approaches addressing the problem are mainly based on supervised learning methods. In this Thesis, first some of freely obtainable datasets with motor or motor-imagery paradigms are collected in a uniform format. Then suitable method implemented for data preprocessing and also a method for onset alignment is proposed. Afterwards, various feature selections available in literature implemented. Lastly, supervised methods are used to verify resulted data and then unsupervised methods are applied for user command detection. Here a method id proposed for making constraints on Gaussian Mixture Model. Also a method for transferring prior knowledge from experiments with different electrode placements proposed and examined. Analyzing results shows acceptable performance for supervised methods and challenges in unsupervised methods. In coclusion possible causes of this results are discussed
  9. Keywords:
  10. Signal Processing ; Unsupervised Learning ; Brain-Computer Interface (BCI) ; Electroencphalogram ; Gaussian Mixture Modeling ; Electroencephalography

 Digital Object List

 Bookmark

...see more